Association of Antiepileptic Drug Usage, Trace Elements and Thyroid Hormone Status

C Zevenbergen, Trace elements and thyroid 174:4 425–432 Clinical Study T I M Korevaar, hormones A Schuette and others

Association of antiepileptic drug usage, trace elements and thyroid hormone status

Chantal Zevenbergen1,2,*, Tim I M Korevaar1,2,*, Andrea Schuette3,*, Robin P Peeters1,2, Marco Medici1,2, Theo J Visser1,2, Lutz Schomburg3 and W Edward Visser1,2

1Department of Internal Medicine and 2Rotterdam Thyroid Center, Erasmus Medical Center, Wytemaweg 80, 3015 Correspondence CN Rotterdam, The Netherlands and 3Institut fu¨ r Experimentelle Endokrinologie, Charite´ -Universita¨ tsmedizin should be addressed Berlin, Augustenburger Platz 1, D-13353 Berlin, Germany to W E Visser *(C Zevenbergen, T I M Korevaar, A Schuette contributed equally to this work) Email [email protected]

Abstract

Background: Levels of thyroid hormone (TH) and trace elements (copper (Cu) and selenium (Se)) are important for development and function of the brain. Anti-epileptic drugs (AEDs) can inﬂuence serum TH and trace element levels. As the relationship between AEDs, THs, and trace elements has not yet been studied directly, we explored these interactions. Method: In total 898 participants, from the Thyroid Origin of Psychomotor Retardation study designed to investigate thyroid parameters in subjects with intellectual disability (ID), had data available on serum Se, Cu, thyroid stimulating hormone

(TSH), free thyroxine (FT4), tri-iodothyronine (T3), reverse T3,T4, and thyroxine-binding globulin (TBG); 401 subjects were on AED treatment. Differences in trace elements according to medication usage was investigated using ANOVA, and associations between trace elements and thyroid parameters were analysed using (non-) linear regression models. Results: Study participants were not deﬁcient in any of the trace elements analyzed. AED (carbamazepine, valproate and phenytoin) usage was negatively associated with serum Se and showed compound-speciﬁc associations with Cu levels.

After correction for drug usage, Se was positively associated with TSH levels, negatively associated with FT4 levels, and

positively with T3 levels. Cu was positively associated with T4,T3, and rT3, which was largely dependent on TBG levels. European Journal of Endocrinology Conclusion: The subjects with ID did not display profound deﬁciencies in trace element levels. AEDs were associated with serum Se and Cu levels, while serum Se and Cu were also associated with thyroid parameters. Further studies on the underlying mechanisms and potential clinical importance are warranted.

European Journal of Endocrinology (2016) 174, 425–432

Introduction

Thyroid hormone (TH) is important for the normal among endocrine disorders (4) at varying prevalences development and energy metabolism of almost all according to the group of patients that is being tissues (1). The thyroid gland pre-dominantly secretes studied (5, 6).

the inactive pro-hormone thyroxine (T4), which is locally Essential trace elements, like Se and copper (Cu), are

converted to the active form tri-iodothyronine (T3)by micronutrients that are present in low concentrations in selenium (Se)-dependent enzymes of the family of the human body and are dependent on a sufﬁciently high

iodothyronine deiodinases (2).T3 mediates its dietary intake. Trace elements are essential for many major effects by binding to nuclear TH receptors (TRs) enzymatic reactions and are therefore important for which regulate the transcription of target genes (3). proper functioning of biochemical pathways and the Abnormalities in TH status are commonly found endocrine system (7).

www.eje-online.org Ñ 2016 European Society of Endocrinology Published by Bioscientiﬁca Ltd. DOI: 10.1530/EJE-15-1081 Printed in Great Britain

Downloaded from Bioscientifica.com at 09/25/2021 03:35:18AM via free access Clinical Study C Zevenbergen, Trace elements and thyroid 174:4 426 T I M Korevaar, hormones A Schuette and others

Sufficient levels of both TH and trace elements are The aim of this study was to analyze the associations important for normal development of the brain (8). between AEDs, Se, Cu, and thyroid parameters using Abnormalities in TH signaling can give rise to neurocog- the TOP-R cohort. nitive impairment, as illustrated by subjects with mutations in the gene encoding the TH transporter monocarboxylate transporter 8 (MCT8) (9, 10) or in the Subjects and methods a TR THRA (encoding TR ) (11, 12, 13, 14). Adequate levels TOP-R study of Cu are also important for normal development of the brain, as Cu is important in energy metabolism, anti- The TOP-R study is a nation-wide cohort study in The oxidative defense, and the production of neurotransmit- Netherlands, which was designed to investigate thyroid ters (15). Also, it was shown that TH or Cu deficiencies parameters in subjects with intellectual disability (ID). resulted in similar defects during rodent cerebral cortical The study population has been described in detail before development (16). The brain appears to preserve Se levels (23). In short, subjects were excluded if the aetiology of the during Se shortage, as rats on a Se deficient diet did not ID was known. Information about other medication was G display any neurological symptoms, because of only unavailable. Mean age was 47.9 0.5 years, and 47% were slightly reduced brain Se levels (17).However,mice women. The study was approved by the medical ethics lacking selenoprotein P (SelP), which is an important Se committee of the Erasmus University Medical Centre. carrier, showed very low brain Se levels together with Written informed consent was obtained from the legally authorized representatives. Of the 946 eligible movement disorders and occasional seizures (18). individuals, data on TH and trace element serum levels Only limited data on the relationship between TH and were available in 898 subjects, and information about Se and Cu are currently available. It was shown in rats that thyroid medication (anti-thyroid drugs and L-T replace- adequate Se in nutrition supports TH synthesis and 4 ment therapy) was available in 806 subjects and about metabolism and protects the thyroid gland from AEDs in 786 subjects. Full case analyses according to this damage by excessive exposure to reactive chemicals (19). data did not change the results (data not shown). After Furthermore, in humans, high levels of Cu were associated exclusion of subjects that used thyroid-interfering medi- with higher levels of both T and T (20). As a corollary, 3 4 cation (nZ40) and/or thyroid peroxidase antibodies Bastian et al. (16, 21, 22) showed in different rat (TPOAb) positive individuals (cut-off O60 IU/ml; nZ24), studies that fetal deficiency of Cu resulted in impaired 834 subjects were included in one or more analyses. After

European Journal of Endocrinology TH-regulated brain gene expression. While these studies exclusion, 372 of 725 eligible patients used AEDs (51.3%). suggested an interaction of the trace elements Se and Cu and TH during brain development, its relevance for humans is unknown. Measurements of thyroid status, Se, and Cu In the Thyroid Origin of Psychomotor Retardation Serum samples were stored frozen at K20 8C. Serum T4, (TOP-R) study, thyroid parameters of subjects with FT4, and thyroid-stimulating hormone (TSH) were intellectual disability (ID) were extensively profiled (23). measured by chemiluminescence assays (Vitros ECI In this cohort, it has been shown that TH profiles in Immunodiagnostic System; Ortho-Clinical Diagnostics, subjects without anti-epileptic drugs (AEDs) were com- Inc., Rochester, NY, USA). T3 was measured using an parable with the general population. However, AEDs were in-house RIA and rT3 using a commercial RIA (Immuno- strongly associated with decreased T4, free T4 (FT4), T3, diagnostic Systems, Scottsdale, AZ, USA). Thyroxine- and rT3 levels which is in agreement with other studies binding globulin (TBG) and TPO antibodies were (24, 25, 26). It has been speculated that the changed determined by immunoassay (Immulite 2000, Siemens, thyroid parameters in patients that use AEDs can be Breda, The Netherlands). Se and Cu concentrations were explained by an influence of AEDs on binding proteins determined by total reflection X-ray fluorescence spectro- (27), a stimulation of hepatic degradation or conjugation scopy (34). The method was validated with a Seronorm of TH (28) or an altered peripheral deiodinase activity (23). standard (Sero AS, Billingstad,Norway).Briefly,all Interestingly, AEDs have been associated with changes in samples were diluted 1:1 in a gallium-containing solvent serum levels of Se and Cu (23, 29, 30, 31, 32, 33). However, for standardization. The analysis was performed in the complex relationship between AEDs, THs, Se, and Cu duplicate, and the results of each sample differed by has never been directly studied. !20% for both Se and Cu. In every measurement run,

www.eje-online.org

Downloaded from Bioscientifica.com at 09/25/2021 03:35:18AM via free access Clinical Study C Zevenbergen, Trace elements and thyroid 174:4 427 T I M Korevaar, hormones A Schuette and others

a human control serum was included, allowing the A P<0.0001 B P=0.0009

calculation of an intra-assay coefﬁcient of variation of 120 120 g/l) g/l) µ 7% (Se) and 8% (Cu) and an inter-assay coefﬁcient of µ variation of 12% (Se) and 18% (Cu) respectively. 110 110

100 100 Selenium in ( Selenium in ( Statistical analysis No Yes No Yes To satisfy model assumptions, TSH levels were logarithmi- Carbamazepine usage Valproate usage cally transformed. The association between trace elements CDP<0.0001 P=0.04 and thyroid parameters was investigated using ordinary 2800 2800 g/l) least squares linear regression models with restricted cubic g/l) µ µ 2600 2600 splines utilizing three to ﬁve knots. Figures show back-

transformed axis values for TSH, and all associations were 2400 2400 Copper in ( adjusted for sex, age, and relevant medication usage Copper in ( (covariates set to mean levels or most appearing category). 2200 2200 The analyses were adjusted for medication usage when a No Yes No Yes Carbamazepine usage Valproate usage drug was associated with a thyroid function parameter in a backward linear regression analysis, utilizing a cut-off of E P=0.0009 2800 P!0.15. In addition, a variable for missing data on AED g/l) usage was added but did not reach the threshold for any µ 2600 thyroid parameters. 2400

We investigated differences in the ratio of T3 to rT3 Copper in (

using linear regression analyses with a product term of the 2200 independent thyroid function parameter and trace No Yes element. Subsequently, to allow for non-linear associa- Phenÿtoïn usage tions/interactions, a sensitivity analysis was performed by adding quadratic terms and/or a product term of the trace Figure 1 element or thyroid function variables; these were main- The association between AED usage and serum Se or Cu and as 2 tained according to P values or changes in R . Interaction predicted mean (gray dot) and 95% CI (gray line). In addition, European Journal of Endocrinology ﬁgures show the associations between thyroid function there was no dose-dependent effect of AED usage for A, B, C, D parameters according to low (red line with 95% CI) or high and E (PZ0.16; PZ0.32; PZ0.57; PZ0.40; and PZ0.62 respect- (blue line with 95% CI) trace element values (rounded ively). All analyses were performed in subjects with AED usage number of 10th percentile or 90th percentile respectively). data available after exclusion of subjects with thyroid All statistical analyses were performed using medication usage or TPOAb positivity and were adjusted for R Statistical Software v 3.03 (1) (package rms or visreg) or sex and age. SPSS version 21.0 for Windows.

(Fig. 1C, D and E). Other AED usage was not associated Results with serum Se or Cu levels, and there was no association Descriptive statistics of the study population are shown between the daily dosage of AEDs and Se, Cu, or TH levels in Supplementary Table 1, see section on supplementary (data not shown). Thus, commonly prescribed AEDs affect data given at the end of this article. First, we studied serum Se and Cu levels in a compound-speciﬁc way. the association between AED usage and serum Se and Cu Next, we studied associations between serum Se and levels (Fig. 1). Cu levels and thyroid parameters. The associations None of the study participants were deﬁcient in Se or between serum Se levels and thyroid parameters are Cu, whether they were on AED treatment or not. shown in Fig. 2. After correction for drug usage, Se was Carbamazepine or valproate usage was associated with positively associated with TSH levels (bGS.E.M; 0.0008G Z lower serum Se levels (Fig. 1A and B). Usage of carbama- 0.0003; P 0.03), negatively associated with FT4 levels zepine or phenytoin was associated with higher serum (linearly for Se below 125: K0.0212G0.0065; PZ0.001),

Cu levels and valproate usage with lower serum Cu levels and positively with T3 levels (linearly for Se below 125:

www.eje-online.org

Downloaded from Bioscientifica.com at 09/25/2021 03:35:18AM via free access Clinical Study C Zevenbergen, Trace elements and thyroid 174:4 428 T I M Korevaar, hormones A Schuette and others

AB sex differences. The association between Cu and TBG 2.00 20 was stronger amongst women, as compared to men (P interactionZ0.0007; Supplementary Figure 2,see 1.72 19 section on supplementary data given at the end of

in (pmol/l) 18 4 this article). After correction for TBG, the positive TSH in (mU/l) 1.46 FT 17 associations between serum Cu and T3,rT3, and T4 levels P=0.03 P=0.009 disappeared (Supplementary Figure 3). Serum Cu levels 50 100 150 200 50 100 150 200 were not associated with changes in the T3/rT3 ratio Se (µg/l) Se (µg/l) (Supplementary Figure 1B). CD

1.7 0.36

1.6

in (nmol/l) Discussion

3 0.34

in (nmol/l) 1.5 3

T TH and the trace elements Se and Cu are important for 0.32 1.4 Reverse T normal neurocognitive development, and abnormal brain P<0.0001 P=0.69 development may increase AED usage. The relationship 50 100 150 200 50 100 150 200 Se (µg/l) Se (µg/l) A B EF 21 1.86 19 105 1.72 18 100 20

1.59 in (pmol/l) 4 in (nmol/l) 4 TSH in (mU/l) FT TBG in (mg/l)

T 17 95 1.46 19 P=0.43 P=0.46 1000 2000 3000 4000 5000 1000 2000 3000 4000 5000 50 100 150 200 50 100 150 200 Cu (µg/l) Cu (µg/l) Se (µg/l) Se (µg/l) C D 1.90 0.40 Figure 2 1.80 in (nmol/l)

Graphs show the association between serum selenium and 3 0.35 1.70 in (nmol/l) European Journal of Endocrinology thyroid function parameters, as predicted mean (black line) and 3 T 0.30 95% CI (grey area). All analyses were performed after exclusion 1.60 Reverse T P=0.002 P=0.0003 of subjects with thyroid medication usage or TPOAb positivity and were adjusted for sex, age, and antiepileptic drug usage 1000 2000 3000 4000 5000 1000 2000 3000 4000 5000 Cu (µg/l) Cu (µg/l) and TBG levels. E F 26 0.0028G0.0005; P!0.001; Fig. 2B and C). Since all three 120 24 deiodinases are selenoproteins, we also investigated 110 22 possible effects of trace elements on peripheral deiodi- in (nmol/l)

4 20 T nase activity. As a proxy for peripheral TH deiodination, 100 TBG in (mg/l) 18 we studied the association between serum Se, Cu, and P<0.0001 P=0.009 T3/rT3 ratio. Serum Se was positively associated with the 1000 2000 3000 4000 5000 1000 2000 3000 4000 5000 Cu (µg/l) Cu (µg/l) T3/rT3 ratio (0.0122G0.0036; P!0.001; Supplementary Figure 1A, see section on supplementary data given at the end of this article). Figure 3 Next, the associations between serum Cu levels and Graphs show the association between serum Cu and thyroid thyroid parameters were studied. Cu levels were positively function parameters, as predicted mean (black line) and 95% CI

associated with T3,rT3, and T4 levels (Fig. 3A, B and C). (grey area). All analyses were performed after exclusion of Also, a particularly strong association was observed subjects with thyroid medication usage or TPOAb positivity and between Cu and TBG (0.0022G0.0002; P!0.001; Fig. 3F). were adjusted for sex, age, and antiepileptic drug usage, As estrogens may inﬂuence TBG levels, we also investigated but not for TBG levels.

www.eje-online.org

Downloaded from Bioscientifica.com at 09/25/2021 03:35:18AM via free access Clinical Study C Zevenbergen, Trace elements and thyroid 174:4 429 T I M Korevaar, hormones A Schuette and others

between TH, Se, Cu, and AED is currently unclear. In this associated with TSH, inversely associated with FT4, and

study, we analyzed this relationship in the TOP-R cohort, positively associated with T3 levels and the T3/rT3 ratio, as in which Se and Cu and thyroid parameters were a proxy for peripheral deiodination (Fig. 2 and Supple- determined. Our analyses indicated that serum Se levels mentary Figure 1). This is also reminiscent of the

were associated with TSH, FT4, and T3 levels, while serum constellation of critical illness, where a Se deﬁcit often

Cu levels were associated with T3,rT3, and T4, via changes parallels a low T3 syndrome. These results are in line with in TBG levels. several studies in rats, in which Se deﬁciency decreased In this cohort of patients with ID, Se levels were liver and kidney D1 activities in combination with modest

signiﬁcantly lower in patients that use valproate or alterations in thyroid parameters (increased T4 and

carbamazepine, whereas phenytoin and carbamazepine decreased T3 levels) (46, 47, 48, 49, 50). However, the usage was associated with increased Cu levels (Fig. 1). majority of studies with human subjects on the inter- Unfortunately, these measurements were performed action of Se and TH yielded inconsistent results (51). during treatment with AEDs; therefore, no data is available Our results indicated that even low normal Se levels may on pre- or post-treatment levels. Many studies described already impact negatively on the TH state. However, net the effects of AEDs on trace element levels like Se and Cu, effects of limiting Se availability for the expression of the although the effects of different classes of AEDs were three deiodinase isoenzymes in different human tissues neither consistent in magnitude nor in direction (29, 30, are difficult to predict, as hierarchical principles control 31, 32, 35, 36, 37). There are several potential reasons for the expression of the selenoenzymes with organ-specific these inconsistencies, ranging from dosage and duration preferences (52). of AED usage, age, health, and nutritional status of the Cu is necessary in many metabolic processes and plays patients to baseline trace element concentrations which an important role in endogenous anti-oxidative defense differ profoundly in different geographical areas as well as mechanisms (53). Studies of Bastian et al. (16, 21, 22) in the small number of subjects studied (38). Due to the mice showed that low levels of Cu were associated with a observational nature of this study, it is very difficult to decreased TH state, which interfered with normal brain draw any conclusions on the causative or mechanistic development. Similarly, serum ceruloplasmin and Cu features of these results. It is very well possible, although levels have recently been proposed as direct biomarkers speculative, that AEDs interfere with the transport, of TH signaling (54). In line with these studies, we metabolism, or excretion of trace elements, as is described observed strong positive associations between serum Cu

for TH (23, 24, 25, 26, 27, 28). Furthermore, although the and T3,rT3, and T4 levels (Fig. 3). However, we were ﬁrst

European Journal of Endocrinology generalizability of our ﬁndings may be limited, it is to describe that these effects are totally driven by TBG

important to be aware of confounders if Se or Cu levels (Supplementary Figure 3). Indeed, FT4 levels are not are analyzed in subjects on commonly prescribed AEDs. affected by Cu state. After correction for TBG, TH Depending on the baseline level of the population, this parameters were not signiﬁcantly associated with serum may cause a physiologically meaningful disbalance. Cu anymore, suggesting that the effects of Cu on thyroid Se is a trace element that is incorporated in seleno- function are mediated via changes in TBG. This may be cysteine, which is required for the normal production of explained by the fact that Cu may directly affect selenoproteins (39, 40, 41). In the last decades, accumu- hepatic TBG expression, secretion, or turnover. TBG is lating evidence has shown that Se plays an essential role in the main transport protein for TH in serum, and many TH biosynthesis and metabolism as well as in normal substances are known to inﬂuence TBG concentrations, thyroid function (42, 43). The thyroid gland contains the such as estrogens, androgens, glucocorticoids, and heroin highest Se concentration among human tissues due to the (55, 56). Most extensively studied are estrogens, which expression of several selenoproteins that are important in increased TBG levels by slowing its clearance via the the maintainance of normal TH metabolism (deiodinases) liver, therefore increasing its half-life (55). In addition, and the protection of thyroid cells against oxidative estrogens have also been shown to increase serum Cu damage such as glutathione peroxidases (43, 44).IfSeis levels, which are most likely driven by an increase limiting, lower levels of selenoproteins are synthesized in ceruloplasmin, the transport protein of Cu (57, 58, 59, which may potentially disturb peroxide-dependent iodi- 60, 61). The association of Cu with TBG was stronger in nation of thyroglobulin in the thyroid, thyrocyte defence women, which is consistent with a positive effect of Cu systems, and TH activation and inactivation (19, 41, 45). on TBG levels via estrogens, although this did not fully In this study, we showed that serum Se was positively explain our observations.

www.eje-online.org

Downloaded from Bioscientifica.com at 09/25/2021 03:35:18AM via free access Clinical Study C Zevenbergen, Trace elements and thyroid 174:4 430 T I M Korevaar, hormones A Schuette and others

Our study has strengths and limitations. The strength 5 Pearce EN. Thyroid dysfunction in perimenopausal and postmeno- of this study is that the TOP-R cohort contains one of pausal women. Menopause International 2007 13 8–13. 6 Peeters RP. Thyroid hormones and aging. Hormones 2008 7 28–35. the largest numbers of subjects with available data on (doi:10.14310/horm.2002.1111035) serum Se, Cu, extensive profiling of TH parameters, and 7 Copius Peereboom JW. General aspects of trace elements and commonly prescribed AEDs. It is important to note that health. Science of the Total Environment 1985 42 1–27. (doi:10.1016/ 0048-9697(85)90003-8) the generalizability of this study is limited, as the patients 8 Bernal J. Thyroid hormones and brain development. Vitamins and in the cohort are all diagnosed with unexplained ID. Hormones 2005 71 95–122. Another potential limitation is the observational nature of 9 Friesema EC, Grueters A, Biebermann H, Krude H, von Moers A, Reeser M, Barrett TG, Mancilla EE, Svensson J, Kester MH et al. the study, which precludes the analysis of causality of the Association between mutations in a thyroid hormone transporter and detected associations. Still, many of our findings itself, and severe X-linked psychomotor retardation. Lancet 2004 364 1435–1437. the direction of causality, are supported by results from (doi:10.1016/S0140-6736(04)17226-7) 10 Dumitrescu AM, Liao XH, Best TB, Brockmann K & Refetoff S. A novel in vivo animal studies. syndrome combining thyroid and neurological abnormalities is Together, our analyses indicate that the commonly associated with mutations in a monocarboxylate transporter gene. prescribed AEDs carbamazepineandvalproateaffect American Journal of Human Genetics 2004 74 168–175. (doi:10.1086/ serum Se and Cu levels. Furthermore, while Se levels 380999) 11 Bochukova E, Schoenmakers N, Agostini M, Schoenmakers E, may partially affect TH signaling via modifying the Rajanayagam O, Keogh JM, Henning E, Reinemund J, Gevers E, expression of deiodinases, the effects of Cu on thyroid Sarri M et al. A mutation in the thyroid hormone receptor a gene. parameters may be primarily driven by its effects on TBG. New England Journal of Medicine 2012 366 243–249. (doi:10.1056/ NEJMoa1110296) Future research is needed to explore the underlying 12 van Mullem A, van Heerebeek R, Chrysis D, Visser E, Medici M, mechanisms of the observed associations in the current Andrikoula M, Tsatsoulis A, Peeters R & Visser TJ. Clinical phenotype study and to investigate to what extent trace elements are and mutant TRa1. New England Journal of Medicine 2012 366 1451–1453. (doi:10.1056/NEJMc1113940) risk factors for the development of thyroid diseases. 13 van Mullem AA, Chrysis D, Eythimiadou A, Chroni E, Tsatsoulis A, de Rijke YB, Visser WE, Visser TJ & Peeters RP. Clinical phenotype of a new type of thyroid hormone resistance caused by a mutation of the TRa1 receptor: consequences of LT4 treatment. Journal of Clinical Supplementary data Endocrinology and Metabolism 2013 98 3029–3038. (doi:10.1210/jc. This is linked to the online version of the paper at http://dx.doi.org/10.1530/ 2013-1050) EJE-15-1081. 14 Moran C, Schoenmakers N, Agostini M, Schoenmakers E, Offiah A, Kydd A, Kahaly G, Mohr-Kahaly S, Rajanayagam O, Lyons G et al. An adult female with resistance to thyroid hormone mediated by defective thyroid hormone receptor a. Journal of Clinical Declaration of interest European Journal of Endocrinology Endocrinology and Metabolism The authors declare that there is no conflict of interest that could be 2013 98 4254–4261. (doi:10.1210/jc. perceived as prejudicing the impartiality of the research reported. 2013-2215) 15 Scheiber IF, Mercer JF & Dringen R. Metabolism and functions of copper in brain. Progress in Neurobiology 2014 116 33–57. (doi:10.1016/ j.pneurobio.2014.01.002) Funding 16 Bastian TW, Prohaska JR, Georgieff MK & Anderson GW. Perinatal iron W E Visser is supported by an Erasmus University Fellowship. L Schomburg and copper deficiencies alter neonatal rat circulating and brain thyroid received support from the Deutsche Forschungsgemeinschaft DFG hormone concentrations. Endocrinology 2010 151 4055–4065. (Scho 849/4-1). A Schuette received a PhD stipend from the Berlin- (doi:10.1210/en.2010-0252) Brandenburg School for Regenerative Therapies (BSRT). 17 Savaskan NE, Brauer AU, Kuhbacher M, Eyupoglu IY, Kyriakopoulos A, Ninnemann O, Behne D & Nitsch R. Selenium deficiency increases susceptibility to glutamate-induced excitotoxicity. FASEB Journal 2003 17 112–114. References 18 Schweizer U, Brauer AU, Kohrle J, Nitsch R & Savaskan NE. Selenium and brain function: a poorly recognized liaison. Brain Research. 1 Oppenheimer JH. Evolving concepts of thyroid hormone action. Brain Research Reviews 2004 45 164–178. (doi:10.1016/j.brainresrev. Biochimie 1999 81 539–543. (doi:10.1016/S0300-9084(99)80107-2) 2004.03.004) 2 Bianco AC & Kim BW. Deiodinases: implications of the local control of 19 Zimmermann MB & Kohrle J. The impact of iron and selenium thyroid hormone action. Journal of Clinical Investigation 2006 116 deficiencies on iodine and thyroid metabolism: biochemistry and 2571–2579. (doi:10.1172/JCI29812) relevance to public health. Thyroid 2002 12 867–878. (doi:10.1089/ 3 Yen PM. Physiological and molecular basis of thyroid hormone action. 105072502761016494) Physiological Reviews 2001 81 1097–1142. 20 Jain RB. Thyroid function and serum copper, selenium, and zinc in 4 Golden SH, Robinson KA, Saldanha I, Anton B & Ladenson PW. Clinical general US population. Biological Trace Element Research 2014 159 Review: Prevalence and incidence of endocrine and metabolic disorders 87–98. (doi:10.1007/s12011-014-9992-9) in the United States: a comprehensive review. Journal of Clinical 21 Bastian TW, Anderson JA, Fretham SJ, Prohaska JR, Georgieff MK & Endocrinology and Metabolism 2009 94 1853–1878. (doi:10.1210/jc. Anderson GW. Fetal and neonatal iron deficiency reduces thyroid 2008-2291) hormone-responsive gene mRNA levels in the neonatal rat

www.eje-online.org

Downloaded from Bioscientifica.com at 09/25/2021 03:35:18AM via free access Clinical Study C Zevenbergen, Trace elements and thyroid 174:4 431 T I M Korevaar, hormones A Schuette and others

hippocampus and cerebral cortex. Endocrinology 2012 153 5668–5680. patients. Journal of Pharmacological Sciences 2004 96 465–473. (doi:10.1210/en.2012-1067) (doi:10.1254/jphs.FPJ04032X) 22 Bastian TW, Prohaska JR, Georgieff MK & Anderson GW. Fetal and 37 Verrotti A, Basciani F, Trotta D, Pomilio MP, Morgese G & Chiarelli F. neonatal iron deficiency exacerbates mild thyroid hormone Serum copper, zinc, selenium, glutathione peroxidase and superoxide insufficiency effects on male thyroid hormone levels and brain thyroid dismutase levels in epileptic children before and after 1 year of sodium hormone-responsive gene expression. Endocrinology 2014 155 valproate and carbamazepine therapy. Epilepsy Research 2002 48 71–75. 1157–1167. (doi:10.1210/en.2013-1571) (doi:10.1016/S0920-1211(01)00322-9) 23 Visser WE, de Rijke YB, van Toor H & Visser TJ. Thyroid status in a large 38 Combs GF Jr. Selenium in global food systems. British Journal of cohort of patients with mental retardation: the TOP-R (Thyroid Nutrition 2001 85 517–547. (doi:10.1079/BJN2000280) Origin of Psychomotor Retardation) study. Clinical Endocrinology 2011 39 Lu J & Holmgren A. Selenoproteins. Journal of Biological Chemistry 2009 75 395–401. (doi:10.1111/j.1365-2265.2011.04089.x) 284 723–727. (doi:10.1074/jbc.R800045200) 24 Liewendahl K, Majuri H & Helenius T. Thyroid function tests in 40 Kohrle J & Gartner R. Selenium and thyroid. Best Practice & Research. patients on long-term treatment with various anticonvulsant drugs. Clinical Endocrinology & Metabolism 2009 23 815–827. (doi:10.1016/ Clinical Endocrinology 1978 8 185–191. (doi:10.1111/j.1365-2265.1978. j.beem.2009.08.002) tb01493.x) 41 Drutel A, Archambeaud F & Caron P. Selenium and the thyroid gland: 25 Verrotti A, Laus M, Scardapane A, Franzoni E & Chiarelli F. Thyroid more good news for clinicians. Clinical Endocrinology 2013 78 155–164. hormones in children with epilepsy during long-term administration (doi:10.1111/cen.12066) of carbamazepine and valproate. European Journal of Endocrinology/ 42 Schomburg L & Kohrle J. On the importance of selenium and iodine European Federation of Endocrine Societies 2009 160 81–86. (doi:10.1530/ metabolism for thyroid hormone biosynthesis and human health. EJE-08-0325) Molecular Nutrition & Food Research 2008 52 1235–1246. (doi:10.1002/ 26 Lossius MI, Tauboll E, Mowinckel P & Gjerstad L. Reversible effects of mnfr.200700465) antiepileptic drugs on thyroid hormones in men and women with 43 Kohrle J, Jakob F, Contempre B & Dumont JE. Selenium, the thyroid, epilepsy: a prospective randomized double-blind withdrawal study. and the endocrine system. Endocrine Reviews 2005 26 944–984. Epilepsy & Behavior 2009 16 64–68. (doi:10.1016/j.yebeh.2009.07.014) (doi:10.1210/er.2001-0034) 27 Surks MI & DeFesi CR. Normal serum free thyroid hormone 44 Schmutzler C, Mentrup B, Schomburg L, Hoang-Vu C, Herzog V & concentrations in patients treated with phenytoin or carbamazepine. Kohrle J. Selenoproteins of the thyroid gland: expression, localization A paradox resolved. Journal of the American Medical Association 1996 275 and possible function of glutathione peroxidase 3. Biological Chemistry 1495–1498. (doi:10.1001/jama.1996.03530430039036) 2007 388 1053–1059. (doi:10.1515/BC.2007.122) 28 Liewendahl K, Helenius T, Majuri H, Ebeling P & Ahlfors UG. Effect of 45 Vanderpas JB, Contempre B, Duale NL, Goossens W, Bebe N, Thorpe R, anticonvulsant and antidepressant drugs on iodothyronines in serum. Ntambue K, Dumont J, Thilly CH & Diplock AT. Iodine and selenium Scandinavian Journal of Clinical and Laboratory Investigation 1980 40 deficiency associated with cretinism in northern Zaire. American 767–774. (doi:10.3109/00365518009095594) Journal of Clinical Nutrition 1990 52 1087–1093. 29 Doneray H, Kara IS, Karakoc A, Tan H & Orbak Z. Serum thyroid 46 Beckett GJ, MacDougall DA, Nicol F & Arthur R. Inhibition of type I and hormone profile and trace elements in children receiving valproic type II iodothyronine deiodinase activity in rat liver, kidney and brain acid therapy: a longitudinal and controlled study. Journal of Trace produced by selenium deficiency. Biochemical Journal 1989 259 Elements in Medicine and Biology 2012 26 243–247. (doi:10.1016/j.jtemb. 887–892. (doi:10.1042/bj2590887) 2012.03.001) 47 Bermano G, Nicol F, Dyer JA, Sunde RA, Beckett GJ, Arthur JR & 30 Kaji M, Ito M, Okuno T, Momoi T, Sasaki H, Yamanaka C, Yorifuji T & Hesketh JE. Tissue-specific regulation of selenoenzyme gene expression Mikawa H. Serum copper and zinc levels in epileptic children with European Journal of Endocrinology during selenium deficiency in rats. Biochemical Journal 1995 311 valproate treatment. Epilepsia 1992 33 555–557. (doi:10.1111/ 425–430. (doi:10.1042/bj3110425) j.1528-1157.1992.tb01709.x) 48 DePalo D, Kinlaw WB, Zhao C, Engelberg-Kulka H & St Germain DL. 31 Karikas GA, Schulpis KH, Bartzeliotou A, Karakonstantakis T, Effect of selenium deficiency on type I 5’-deiodinase. Journal of Biological Georgala S, Kanavaki I, Demetriou E & Papassotiriou I. Lipids, Chemistry 1994 269 16223–16228. lipoproteins, apolipoproteins, selected trace elements and minerals in the serum of children on valproic acid monotherapy. Basic & Clinical 49 Chanoine JP, Safran M, Farwell AP, Tranter P, Ekenbarger DM, Pharmacology & Toxicology 2006 98 599–603. (doi:10.1111/j.1742-7843. Dubord S, Alex S, Arthur JR, Beckett GJ, Braverman LE et al. Selenium 2006.pto_402.x) deficiency and type II 5’-deiodinase regulation in the euthyroid and 32 Kurekci AE, Alpay F, Tanindi S, Gokcay E, Ozcan O, Akin R, Isimer A & hypothyroid rat: evidence of a direct effect of thyroxine. Endocrinology Sayal A. Plasma trace element, plasma glutathione peroxidase, and 1992 131 479–484. superoxide dismutase levels in epileptic children receiving antiepileptic 50 Meinhold H, Campos-Barros A, Walzog B, Kohler R, Muller F & drug therapy. Epilepsia 1995 36 600–604. (doi:10.1111/j.1528-1157. Behne D. Effects of selenium and iodine deficiency on type I, type II and 1995.tb02574.x) type III iodothyronine deiodinases and circulating thyroid hormones 33 Armutcu F, Ozerol E, Gurel A, Kanter M, Vural H, Yakinci C & Akyol O. in the rat. Experimental and Clinical Endocrinology 1993 101 87–93. Effect of long-term therapy with sodium valproate on nail and serum (doi:10.1055/s-0029-1211212) trace element status in epileptic children. Biological Trace Element 51 Hess SY. The impact of common micronutrient deficiencies on iodine Research 2004 102 1–10. (doi:10.1385/BTER:102:1-3:001) and thyroid metabolism: the evidence from human studies. Best Practice 34 Hoeflich J, Hollenbach B, Behrends T, Hoeg A, Stosnach H & & Research. Clinical Endocrinology & Metabolism 2010 24 117–132. Schomburg L. The choice of biomarkers determines the selenium status (doi:10.1016/j.beem.2009.08.012) in young German vegans and vegetarians. British Journal of Nutrition 52 Schomburg L & Schweizer U. Hierarchical regulation of 2010 104 1601–1604. (doi:10.1017/S0007114510002618) selenoprotein expression and sex-specific effects of selenium. 35 Sarangi SC, Tripathi M, Kakkar AK & Gupta YK. Effect of antiepileptic Biochimica et Biophysica Acta 2009 1790 1453–1462. (doi:10.1016/ therapy on trace elements status in Indian population in a tertiary care j.bbagen.2009.03.015) hospital from northern India: a cross sectional study. Epilepsy Research 53 Alturfan AA, Zengin E, Dariyerli N, Alturfan EE, Gumustas MK, 2014 108 917–927. (doi:10.1016/j.eplepsyres.2014.01.014) Aytac E, Aslan M, Balkis N, Aksu A, Yigit G et al. Investigation of zinc 36 Hamed SA, Abdellah MM & El-Melegy N. Blood levels of trace elements, and copper levels in methimazole-induced hypothyroidism: relation electrolytes, and oxidative stress/antioxidant systems in epileptic with the oxidant–antioxidant status. Folia Biologica 2007 53 183–188.

www.eje-online.org

Downloaded from Bioscientifica.com at 09/25/2021 03:35:18AM via free access Clinical Study C Zevenbergen, Trace elements and thyroid 174:4 432 T I M Korevaar, hormones A Schuette and others

54 Mittag J, Behrends T, Nordstrom K, Anselmo J, Vennstrom B & 58 Buchwald A. Serum copper elevation from estrogen effect, masquer- Schomburg L. Serum copper as a novel biomarker for resistance to ading as fungicide toxicity. Journal of Medical Toxicology 2008 4 30–32. thyroid hormone. Biochemical Journal 2012 443 103–109. (doi:10.1042/ (doi:10.1007/BF03160948) BJ20111817) 59 Arredondo M, Nunez H, Lopez G, Pizarro F, Ayala M & Araya M. 55 Tahboub R & Arafah BM. Sex steroids and the thyroid. Best Practice & Inﬂuence of estrogens on copper indicators: in vivo and in vitro studies. Research. Clinical Endocrinology & Metabolism 2009 23 769–780. Biological Trace Element Research 2010 134 252–264. (doi:10.1007/ (doi:10.1016/j.beem.2009.06.005) s12011-009-8475-x) 56 Surks MI & Sievert R. Drugs and thyroid function. New England 60 Hinks LJ, Clayton BE & Lloyd RS. Zinc and copper concentrations in Journal of Medicine 1995 333 1688–1694. (doi:10.1056/ leucocytes and erythrocytes in healthy adults and the effect of oral NEJM199512213332507) contraceptives. Journal of Clinical Pathology 1983 36 1016–1021. 57 Sobbrio GA, Granata A, Granese D, D’Arrigo F, Panacea A, Nicita R, (doi:10.1136/jcp.36.9.1016) Pulle C & Trimarchi F. Sex hormone binding globulin, cortisol binding 61 Raymunt J & Russ EM. Inﬂuence of estrogens on total serum globulin, thyroxine binding globulin, ceruloplasmin: changes in copper and caeruloplasmin. Proceedings of the Society for Experimental treatment with two oral contraceptives low in oestrogen. Clinical and Biology and Medicine 1956 92 465–466. (doi:10.3181/00379727-92- Experimental Obstetrics & Gynecology 1991 18 43–45. 22512)

Received 4 November 2015 Revised version received 17 December 2015 Accepted 23 December 2015 European Journal of Endocrinology

www.eje-online.org

Downloaded from Bioscientifica.com at 09/25/2021 03:35:18AM via free access